|
| 基于注意力机制及贝叶斯优化的圆锥滚子轴承故障状态监测 |
| Fault Condition Monitoring of Tapered Roller Bearings Based on Attention Mechanism and Bayesian Optimization |
| 投稿时间：2025-04-27 修订日期：2025-05-13 |
| DOI： |
| 中文关键词: 故障诊断 CNN 注意力机制 贝叶斯优化 |
| 英文关键词: Fault diagnosis CNN Attention mechanism Bayesian optimization |
| 基金项目:国家自然科学基金项目（51905001）；安徽未来技术研究院企业合作项目（2023qyhz22）；安徽工程大学校级项目（Xjky2022012） |
|
| 摘要点击次数: 301 |
| 全文下载次数: 0 |
| 中文摘要: |
| 针对贝叶斯优化准确率低和处理时间长的问题。提出一种基于注意力机制和贝叶斯优化共同作用的模型。注意力机制可以通过学习到的权重，有针对性地关注不同位置的信息，从而更好地捕捉长距离的依赖关系，并提升模型在序列任务上的性能，但是这也增加了参数的数量，而贝叶斯优化可以指导搜索方向，可以有效地避免了尝试大量无效的参数组合，减少参数处理的时间。最终实验证明：该网络结构较之前的结构拥有更好的性能。 |
| 英文摘要: |
| To address the issues of low accuracy and long processing time. a model based on the combined action of attention mechanism and Bayesian optimization is proposed. The attention mechanism can focus on information at different positions through learned weights, thereby better capturing long-range dependencies and enhancing the performance of the model in sequential tasks. However, this also increases the number of parameters. Bayesian optimization, on the other hand, can guide the search direction and effectively avoid attempting a large number of invalid parameter combinations, reducing parameter processing time. Finally, the experiment proves that this network structure exhibits superior performance compared to previous structures. |
|
View Fulltext
查看/发表评论 下载PDF阅读器 |
| 关闭 |
分享按钮